Ameiva fuscata on , : natural history and interactions with Anolis oculatus SETH M. RUDMAN1, ROBERT POWELL2,4 and JOHN S. PARMERLEE, JR.3

1 Department of Biology, University of Rochester, Rochester, NY 14627, USA. 2 Department of Biology, Avila University, Kansas City, MO 64145, USA. 3 Department of Biology, Johnson County Community College, Overland Park, KS 66210, USA.

4 Corresponding author: [email protected]

ABSTRACT — In June 2008, we studied the natural history of Ameiva fuscata and interactions with sympatric Anolis oculatus in five adjacent habitats on the western (leeward) coast of Dominica, Lesser Antilles. Ameiva activity was positively correlated with mean temperature, peak activity corresponded to peak daily temperatures and we observed greatly reduced or no activity during overcast and rainy periods. Population densities in the five habitats were 138–344 lizards/ha, with the higher densities in areas with the deepest leaf litter and densest canopy cover. Ameiva fuscata is an active forager, with 100% of foraging attempts made while moving (either walking or actively digging or rooting in litter). Lizards foraged together in groups and the groups never entered each other’s activity areas, suggesting the possibility of community structure. Ameiva fuscata is non-territorial, but chases among adult males appeared to be triggered by intersecting paths. Anoles spent significantly less time on the ground when Ameiva fuscata was active, presumably to avoid predation. est Indian islands support abundant and interactions between Ameiva and Anolis. Meier Wdiverse lizard communities. Common & Noble (1991, Desecheo), Fobes et al. (1992, elements of many of these communities include Hispaniola) and Eaton et al. (2002, Anguilla) an active foraging terrestrial lizard in the genus suggested that anoles decrease time spent on the Ameiva and arboreal sit-and-wait ambush foragers ground in the presence of Ameiva. Simmons et in the genus Anolis (Simmons et al., 2005). al. (2005) examined interactions between Ameiva Although West Indian anoline lizards are arguably ameiva and two species of Anolis (A. aeneus and the most intensely studied in the world A. richardii) on Grenada and concluded that both (Losos, 2009), relatively little is known about species of Anolis spend less time on the ground most species of West Indian Ameiva, especially when A. ameiva is active. Kolbe et al. (2008) considering their visibility on the islands where showed a niche shift by Anolis wattsi to lower they occur (Henderson & Powell, 2009). perches and more terrestrial activity in the absence Ameiva are typically heliothermic, active of Ameiva griswoldi on Antiguan satellite islands. foragers, have relatively short activity periods, are From 4–23 June 2008, we conducted a study not territorial and have overlapping home ranges on Dominica to describe activity, population (e.g., Hillman, 1969; Regal, 1978, 1983; Schell densities, foraging behaviour and movements et al., 1993; Vitt & Colli, 1994; Simmons et al., of Ameiva fuscata and to test the hypothesis that 2005). Ameiva fuscata (Fig. 1; and front cover) is Anolis oculatus (Fig. 2) would spend less time on endemic to Dominica and occurs throughout the the ground when A. fuscata is present and active. island in coastal woodlands, generally at elevations below 200 m, but occasionally higher in cultivated METHODS AND MATERIALS areas (Bullock & Evans, 1990; Malhotra & Our study site was at the mouth of the Batali Thorpe, 1999). River on the leeward (western) coast of Dominica Despite the number of islands on which both (N 15° 27.12', W 061° 26.76'). The site was genera are represented, few studies have addressed characterised by habitats including beach,

Herpetological Bulletin [2009] - Number 109 17 Ameiva fuscata, natural history on Dominica

Area Size (m2) insolation Most Abundant Maximum estimated Vegetation number Density (#/ha)

1 1134 mixed ground scrub 39 344 2 2035 sunny/mixed Snake Plant 28 138 3 2554 mixed Mango trees 62 243 4 873 shade/mixed scrub, trees 15 172 5 1925 shade/mixed Coconut trees 57 296

Table 1. Characterisation of areas/habitats (see text) and maximum numbers of Ameiva fuscata observed during any one survey. beachside scrub with Sea Grape (Coccoloba manner designed to avoid multiple encounters uvifera) and dry forest interspersed with Mango with the same lizards, we counted the number of trees (Mangifera sp.). Cleared paths passed individual Ameiva fuscata observed in each of the between and extended into most areas. Open five contiguous habitats and calculated estimated areas were characterised by grasses, herbaceous population densities (numbers of /ha) using forbs and stands of ornamental vegetation, the maximum number of lizards in any one area including extensive areas planted in “Snake Plant” at one time relative to the areas (m2) sampled. A (Sanseveria sp.). data logger (HOBO® TidbiT® v2 Submersible The five habitat areas (Table 1) were identified Temperature Logger, Onset Computer Corp., after observing what appeared to be natural Bourne, Massachusetts, USA) was placed in leaf groupings of Ameiva associated with areas of high litter at site 5 to record hourly temperatures during population densities and corresponding roughly the extent of the study. to distinct habitat types. Areas/habitats were We conducted focal studies on A. characterised by: (1) mix of native and orchard fuscata. Observations ranged from < 1 min to 20 trees with a canopy coverage of 50–80% and a mins in duration. In addition to recording anecdotal sparse 2–3 m high understory, canopy height of observations of individuals or groups of lizards, we 5–15 m with litter depth of 2–4 cm and occasional quantified behaviours using methods of Cooper et human traffic; (2) beachside vegetation with 1–2 al. (2001), recording time spent moving, number m-high understory, canopy coverage of 20–50%, of moves and number of feeding attempts while canopy height 4–12 m, sandy substrate with some stationary or mobile. We used these observations to leaf litter and regular human traffic; (3) orchard calculate moves per minute (MPM), percent time trees with sparse 2–4 m-high understory in some spent moving (PTM) and proportion of feeding areas, canopy coverage ca. 90% over part of the attempts while moving (PAM). area and ca. 10% over the remainder, canopy height Using a paint gun (Forestry Suppliers, Inc., 5–20 m with litter depth of 1–2 cm, abundant fallen Jackson, Mississippi, USA) and two colours of latex fruit and occasional human traffic; (4) mix of native paint diluted 1:1 with water, we marked individual and orchard trees with canopy coverage of 60– A. fuscata active in areas 2 and 5 (along the beach 90% and a dense 1–2 m-high understory, canopy and in a densely shaded area among large boulders), height of 5–20 m with a litter depth of 1–2 cm and attempting to paint all individuals in three sessions occasional human traffic; and (5) orchard trees at each location. The two sites were only ca. 40 m with canopy coverage of 70–90%, leaf litter depth apart and no evident barriers precluded movement 2–4 cm, abundant fallen fruit, sparse understory between the areas. We then observed movements and little human traffic. and behaviours of marked lizards to determine if We conducted 17 45-min surveys at various lizards foraging in one area comprised a grouping times from 0700–1730 h, recording time, extent distinct from those in the other. of cloud cover and any precipitation during each In the same five adjacent habitats, at various times visit. Systematically covering each area in a of day and under varying weather conditions, we

18 Number 109 - Herpetological Bulletin [2009] Ameiva fuscata, natural history on Dominica

Figure 1. Adult male Ameiva fuscata. Photograph by Robert Powell.

Figure 2. Adult male Anolis oculatus. Photograph by Robert Powell.

Herpetological Bulletin [2009] - Number 109 19 Ameiva fuscata, natural history on Dominica conducted 43 10-min surveys of anoles, recording Temperatures were highest just as activity began to lizard position as either on the ground or on elevated decline. The maximum observed population density perches (e.g. vegetation or rocks). Anoles were not in any one area was 344 individuals/ha (Table 1). marked and we may have encountered the same That with the highest density had the deepest leaf individuals more than once. However, we contend litter and the densest canopy cover. Mean MPM for that perches at different times and on different days all A. fuscata was 2.1 ± 0.2 (0.2–4.9). Mean PTM are independent events predicated by either the was 43.8 ± 3.3% (3.0–98.2%). PAM was 100%, presence or absence of Ameiva and that we are not with 55.6% of feeding attempts occurring while guilty of pseudoreplication of data. We recorded animals were searching and 44.3% while in one time, temperature and general weather conditions location but actively digging. during each survey and noted whether A. fuscata Painted individuals were observed daily. The was abundant and active (1), scarce (2), or absent (3). two “subpopulations” were never observed to

Figure 3. Mean percent of Ameiva fuscata active in all areas and mean ambient surface temperatures by time of day.

We used StatView 5.0 (SAS Institute, co-mingle and never found in areas other than Cary, North Carolina, USA) for statistical where they were initially painted. analyses. Means are presented ± one SE. For all We frequently observed individuals of all sizes tests, alpha = 0.05. foraging together with few interactions between lizards. However, large males regularly chased RESULTS one another. These chases would sometimes occur Ameiva activity (percent of the maximum number between individuals that had been foraging close of Ameiva observed in a given area) was positively together for several minutes. In most instances, no correlated with mean temperature (Spearman obvious cause could be discerned. However, based Rank Correlation, Z = 2.46, P = 0.01; Fig. 3), on five observations, the angle at which paths of showing clearly that a single peak activity period moving individuals intersected appeared to trigger corresponded with peak daily temperatures. We at least some chases (i.e., when a large male was observed greatly reduced or no activity during about to intersect the path of another large male). overcast and rainy periods, even at times when We observed 618 anoles during 43 10-min Ameiva were usually active. Observed activity surveys (289 on the ground, 329 on elevated periods were from 08:00–16:00 at a mean surface perches). The mean percent of anoles observed on temperature of 30.1 ± 0.7°C (26.4–33.3°C), the ground during all survey periods was 44.7 ± with a peak during late morning (09:30–11:30). 4.4% (Fig. 4), but this percent varied considerably

20 Number 109 - Herpetological Bulletin [2009] Ameiva fuscata, natural history on Dominica

Figure 4. Mean percent of Anolis oculatus observed on the ground throughout the day. throughout the day (Fig. 5). Significantly fewer A. fuscata usually is active (the large number on anoles (12.2 ± 1.7%) ventured onto the ground ground at 11:00 in Fig. 4 reflects observations on when Ameiva fuscata was abundant and active rainy days). than when few individuals were active (21.9 ± 6.7%) or when no Ameiva (66.9 ± 2.5%) were seen DISCUSSION (one-way contingency tests, all χ2 ≥ 150.4, all P ≤ Previous studies conducted at the same time of 0.0001). Anoles were found on the ground in high year have shown that some populations of Ameiva numbers at times of day when Ameiva were not (A. chrysolaema, Schell et al., 1993; A. ameiva, active (07:00–08:00 and 15:00–17:00), but also Simmons et al., 2005) exhibit two activity peaks, during rainy/overcast periods, even at times when but that was not evident in this population of A.

Figure 5. Mean number of Anolis oculatus observed on the ground (dark grey) and on elevated perches (light grey) per 10-min survey at different levels of Ameiva fuscata activity (abundant and active = 1, scarce = 2, or absent = 3).

Herpetological Bulletin [2009] - Number 109 21 Ameiva fuscata, natural history on Dominica fuscata. Observed activity periods and the maximum particular (e.g., Hodge et al., 2003). observed population density corresponded with Home ranges of West Indian species of activity periods and the maximum density (379/ Ameiva are known to overlap (e.g., Kerr et al., ha) observed by Bullock & Evans (1990) for A. 2005; Simmons et al., 2005) and appear to vary fuscata. Although our estimates of density are considerably in size. Simmons et al. (2005) recorded undoubtedly conservative (as they are based on the mean home range sizes for A. ameiva on Grenada maximum number of lizards seen at only one time), of 648 ± 252 m2 (adult males) and 204 ± 55 m2 they are higher than most other population density (adult females). Censky (1995) noted larger home estimates for West Indian species of Ameiva. Meier ranges for A. plei on Anguilla (1551 ± 566 m2 for et al. (1993) calculated 73.7–81.1 A. polops/ha on males, 864 ± 504 m2 for females). However, those Green Cay (U.S. Virgin Islands); Schell et al. (1993) observed by Lewis & Saliva (1987) for A. exsul on presented an estimate of 136–144 A. chrysolaema/ Puerto Rico (377 m2, 174 m2) were slightly smaller ha in an altered habitat in the Dominican Republic; and those reported by Schell et al. (1993) for A. Censky (1996) found 74 A. pleii/ha on Anguilla chrysolaema in a partially confined habitat in the and 91/ha on Dog Island; McNair (2003) presented Dominican Republic (250 ± 122 m2, 115 ± 102 crude and ecological densities of 25 and 128 A. m2), by Meier et al. (1993) for A. polops on Green polops/ha, respectively, on Protestant Cay (U.S. Cay off St. Croix (190 ± 86 m2 for both sexes) Virgin Islands); and R. Powell & R.W. Henderson and by Kerr et al. (2005) for A. erythrocephala (unpubl. data) estimated 33.8–52.2 A. exsul/ on St. Eustatius (101 ± 36 m2, 54 ± 12 m2) were ha on Guana Island (British Virgin Islands). considerably smaller. The consistency with which Only Simmons et al. (2005) calculated a higher marked A. fuscata were resighted only in those density, 460 adult A. ameiva/ha in an “activity areas where originally seen during the current study area” on Grenada. Because population sizes were suggests that groups of individuals have relatively estimated using different methods, however, direct small home ranges where population densities are comparisons may not be appropriate. high. That individuals with the same paint colour Maximum observed densities of A. fuscata were consistently observed together also was presumably reflect the availability of resources in suggestive of a previously undescribed community each area. Those with the highest density (Table structure. In addition, individuals were frequently 1) had the deepest leaf litter and the densest observed making foreleg movements that appeared canopy cover. Abundant leaf litter may harbour to serve a communicative function and seemed more abundant prey. The high tolerance of shaded to be directed at other lizards within the group. habitats is unusual in West Indian species of Ameiva, Furthermore, individuals often interrupted foraging which generally are associated with open situations behaviour and appeared to monitor the activity of (Henderson & Powell, 2009). Historically, however, nearby animals. open habitats may have been present on Dominica Ameiva fuscata and congeners in general have only intermittently after hurricanes, forcing the been described as non-territorial (e.g. Schell et species to adapt to areas with closed canopies, in al., 1993; Hodge et al., 2003) and observations of which they exploit small sun-lit patches to bask. individuals of all sizes foraging together supported Observed values for MPM, PTM and PAM are that contention. However, large males occasionally comparable to those observed by Simmons et al. chased each other while foraging, which would (2005) for A. ameiva (MPM = 5.6 ± 0.5, PTM = indicate that defence of individual space may occur. 51.2 ± 3.9%, PAM= 96.0 ± 2.0%). Cooper et al. This aggression appears to be triggered by the angle (2001) described PAM as a means of quantifying at which paths of actively foraging individuals foraging methods in conjunction with MPM and intersect. Mate-guarding, which has been described PTM. Our data clearly indicated that A. fuscata for some populations of A. pleii (Censky, 1995), employed an active foraging strategy, which is also might result in agonistic interactions, but we characteristic of teiid lizards in general (Cooper observed only one instance of what might have et al., 2001) and West Indian species of Ameiva in been construed as mate-guarding.

22 Number 109 - Herpetological Bulletin [2009] Ameiva fuscata, natural history on Dominica

At our study site, Anolis oculatus was active REFERENCES from dawn to dusk (N.J. Vélez Espinet & E.A. Bullock, D.J. & Evans, P.G.H. (1990). The Daniells, unpubl. data). Time spent on the ground distribution, density and biomass of terrestrial by anoles was not dependent on the time of day reptiles in Dominica, West Indies. J. Zool. or any single environmental condition, but instead (Lond.) 222, 421–443. appeared to be influenced primarily by the presence Censky, E.J. (1995). Mating strategy and of Ameiva. As proposed by Simmons et al. (2005) reproductive success in the teiid lizard, Ameiva and supported by our observations of successful plei. Behavior 132, 529–557. predation on anoles by Ameiva, the principal cause Censky, E.J. (1996). The evolution of sexual size of this behaviour by anoles appears to be predator dimorphism in the teiid lizard Ameiva plei: A avoidance. test of alternative hypotheses. In: Contributions In general, our observations suggest that the to West Indian : A Tribute to Albert natural history of Dominican Ameiva fuscata is Schwartz, pp. 277–289. Powell, R. & Henderson, generally comparable to that of other West Indian R.W. (Eds.), Contributions to Herpetology, Vol. congeners. These lizards are largely terrestrial, 12. Ithaca, New York: Society for the Study heat-loving, non-territorial active foragers. At Amphibians and Reptiles. our study site, however, they did not exhibit Cooper, W.E., Jr., Vitt, L.J., Caldwell, J.P., & Fox, the bimodal activity period evident in at least S.F. (2001). Foraging modes of some American some populations of some congeners and their lizards: Relationships among measurement predilection for foraging in heavily shaded areas variables and discreteness of modes. was unusual. Observations of possible community Herpetologica 57, 65–76. structure and that aggression between adult males Eaton, J.M., Larimer, S.C., Howard, K.G., Powell, might be triggered by intersecting trajectories of R., & Parmerlee, J.S., Jr. (2002). Population moving lizards are new. Although these phenomena densities and ecological release of a solitary currently are supported solely by anecdotal data, species: Anolis gingivinus on Anguilla, West they are worthy of further investigation. Also, Indies. Carib. J. Sci. 38, 27–36. despite warnings by Micco et al. (1997) that Fobes, T.M., Powell, R., Parmerlee, J.S., Jr., information pertaining to a single population might Lathrop, A., & Smith, D.D. (1992). Natural not be extrapolated accurately to other populations history of Anolis cybotes (Sauria: Polychridae) (much less to other species), responses by Anolis in a disturbed habitat in Barahona, Dominican oculatus to Ameiva fuscata were essentially Republic. Carib. J. Sci. 28, 200–207. similar to those observed between anoles and Henderson, R.W. & Powell, R. (2009). Natural Ameiva ameiva on the island of Grenada (Simmons History of West Indian Reptiles and et al., 2005). Amphibians. Gainesville, Univ. Press of Florida. ACKNOWLEDGEMENTS Hillman, P.E. (1969). Habitat specificity in three Arlington James, Forest Officer, Forestry, Wildlife sympatric species of Ameiva (Reptilia; Teiidae). and Parks Division, Ministry of Agriculture & Ecology 50, 476–481. the Environment, Commonwealth of Dominica, Hodge, K.V.D., Censky, E.J., & Powell, R. (2003). was instrumental in issuing permits to conduct The Reptiles and Amphibians of Anguilla, research in Dominica and facilitated our efforts British West Indies. The Valley, Anguilla in myriad ways. Protocols were approved by the National Trust. 72 pp. Avila University Animal Care and Use Committee. Kerr, A.M., Powell, R., & Parmerlee, J.S. Jr. Robert W. Henderson reviewed an earlier draft (2005). Ameiva erythrocephala (Teiidae) on Sint of this manuscript and contributed exceedingly Eustatius, Netherlands Antilles: Baseline data helpful suggestions. Fieldwork was funded by a on a small population in a severely altered grant from the National Science Foundation (USA) habitat. Carib. J. Sci. 41, 162–169. to Robert Powell (DBI-0242589). Kolbe, J.J., Colbert, P.L., & Smith, B.E. (2008).

Herpetological Bulletin [2009] - Number 109 23 Ameiva fuscata, natural history on Dominica

Niche relationships and interspecific interactions populations. Herpetol. Nat. Hist. 5, 147–156. in Antiguan lizard communities. Copeia 2008, Regal, P.J. (1978). Behavioural differences between 261–272. reptiles and mammals: An analysis of activity Lewis, A.R. & Saliva, J.F. (1987). Effects of sex and mental capabilities. In: Behaviour and and size on home range, dominance and activity Neurology of Lizards: An Interdisciplinary budgets in Ameiva exsul (Lacertilia: Teiidae). Colloquium, pp. 183–202. Greenberg, N., Herpetologica 43, 374–383. MacLean, P.D. (Eds.). Rockville, Maryland, Losos, J.B. (2009). Lizards in an Evolutionary Nat. Inst. Mental Health. Tree: Ecology and of Anoles. Regal, P.J. (1983). The adaptive zone and behaviour Berkeley, Univ. California Press. xx + 507 pp. of lizards. In: Lizard Ecology: Studies of a Malhotra, A. & Thorpe, R.S. (1999). Reptiles & Model Organism, pp. 105–118. Huey, R.B., Amphibians of the Eastern . London, Pianka, E., Schoener, T.W. (Eds.). Cambridge, Macmillan Educ. Ltd. ix + 134 pp. Massachusetts, Harvard Univ. Press. Meier, A.J. & Noble, R.E. (1991). Notes on the Schell, P.T., Powell, R., Parmerlee, J.S., Jr., natural history of Anolis desechensis. Florida Lathrop, A., & Smith, D.D. (1993). Notes on the Fld. Nat. 19, 17–18. natural history of Ameiva chrysolaema (Sauria: Meier, A.J., Noble, R.E., & Rathbun, S.L. (1993). Teiidae) from Barahona, Dominican Republic. Population status and notes on the biology of the Copeia 1993, 859–862. St. Croix Ground Lizard on Green Cay (St. Simmons, P.M., Greene, B.T., Williamson, K.E., Croix, U. S. Virgin Islands). Carib. J. Sci. 29, Powell, R., & Parmerlee, J.S., Jr. (2005). 147–152. Ecological interactions within a lizard community Micco, S.M., Lahey, G.J., Sosa, R.A., Powell, R., on Grenada. Herpetologica 61, 124–134. Censky, E.J., & Parmerlee, J.S., Jr. (1997). Vitt, L.J. & Colli, G.R. (1994). Geographical Natural history of Leiocephalus barahonensis ecology of a Neotropical lizard: Ameiva (Tropiduridae) on the Península de ameiva (Teiidae) in Brazil. Can. J. Zool. 72, Barahona, Hispaniola: An examination of two 1986–2008.

24 Number 109 - Herpetological Bulletin [2009]